Autism is a complex and pervasive disorder with often devastating effects on social, cognitive, and language development. The presence of mental retardation and lack of speech are associated with very poor outcomes. The verbal level that children attain during preschool years is a strong predictor of adult adaptive skill and educational levels. Intensive treatment for autism early in life can improve the development of useful speech and decrease the severity of mental retardation. Thanks to the development of better diagnostic tools and a greater level of professional education, autism is being identified in two year olds and in even younger children, with the rationale that the earlier intervention begins, the better the outcomes may be. However, there are no published outcome data on intervention models or effectiveness for children who begin treatment by or before 24 months. Furthermore, some teaching procedures considered appropriate for older children are considered developmentally inappropriate for toddlers. Drs. Dawson and Rogers have implemented a feasibility study of a treatment designed for toddlers with autism using a randomized controlled design. The manualized approach involves a relationship-based frame to accomplish developmentally based objectives using naturalistic application of applied behavior analytic principles and is delivered 1:1 for 25 or more hours per week to 24 toddlers with autism for a two year period. The contrast group receives standard community based intervention. Preliminary results show large and significant group effects after only 12 months and considerable variability of treatment outcomes in both groups. Specific Aim 1 will conduct a multi-site intent-to-treat randomized control trial of early intensive behavioral intervention (Early Start Denver Model - ESDM) compared to standard community treatment involving 108 toddlers with autism (18-24 months of age), 36 at each of three sites, to evaluate the efficacy of very early intervention for improving child outcomes after 12 and 24 months of treatment;Specific Aim 2 will examine predictors, mediators, and moderators of outcomes for children in both groups, with variables involving social environmental risks and protective features, individual child risk or protective features, and biological risks and protective features. Overall, the study is based on Hypothesis 1: ESDM treatment will result in significantly greater improvements in core autism symptoms and developmental rates of progress in multiple areas than community treatment;and Hypothesis 2: Specific individual child variables, number of biological risks, number of social-environmental risks and opportunities, and parent interaction styles will facilitate or modify child outcomes.